基于类向心度的模糊支持向量机

A novel fuzzy support vector machine based on the class centripetal degree

传统支持向量机(SVM)训练含有噪声或野值点的数据时,容易产生过拟合,而模糊支持向量机可以有效地处理这种问题。针对使用样本与类中心之间的距离关系来构建模糊支持向量机隶属度函数的不足,提出了一种基于类向心度的模糊支持向量机(CCD-FSVM)。该方法不仅考虑到样本与类中心之间的关系,还考虑到类中各个样本之间的联系,并用类向心度来表示。将类向心度应用于模糊隶属度函数的设计,能够很好地将有效样本与噪声、野值点样本区分开来,而且可以通过向心度的大小,对混合度比较高的样本进行区分,从而达到提高分类精度的效果。实验结果表明,基于类向心度的模糊支持向量机其分类正确率比支持向量机高,在使用三种不同隶属度函数的FSVM中,该方法的抗噪性能最好,分类性能最强。

The traditional support vector machine （SVM） often falls into over-fitting when outliers are contained in the training data.The fuzzy support vector machine can effectively deal with this prob lem.According to the deficiency of the membership function designed based on the distance between a sample and its cluster center,a novel fuzzy support vector machine based on the class centripetal degree （CCD FSVM） is proposed.It combines the distance between a sample and its cluster center with the re lationship between samples expressed as the class centripetal degree.This function can effectively sepa rate the valid samples from the noises or outliers.Besides,the size of the class centripetal degree can re flect the samples mixed degree.Experimental results show that the fuzzy support vector machine based on the class centripetal degree is more robust than the traditional support vector machine,and it outper forms the other two FSVM counterparts with different membership functions in terms of antinoise and classification performance.